Prosthetic joint infection (PJI) is increasing in incidence and poses diagnostic and therapeutic challenges. One quarter to one fifth of cases currently classified as PJI are microbiologically-negative using classic techniques and even in cases in which microbes are detected, our recent data shows that there are subpopulations of organisms present which are undetected using existing strategies. Further, differentiating infection from aseptic implant failure can be problematic. {Outside of mechanical reasons such as implant malposition, the cause of so-called aseptic implant failure is not well-defined and may be due by unrecognized infection with organisms that are nontraditional causes of PJI.} Microorganisms associated with PJI are found in biofilms on the surface of the implant. We have developed a method that uses vortexing and sonication to sample the implant surface and have consistently shown that biofilm-containing material dislodged from the implant surface provides the ideal specimen for microbial detection, better than periprosthetic tissues and synovial fluid. Nevertheless, there remain many cases of culture-negative PJI. To address this issue, with support of R01 AR056647, we developed a panel of PCR assays targeting, at the genus-/group-level, the most common bacteria known to cause PJI. The sensitivity of the PCR panel was 77% whereas that of culture was 73% when performed on dislodged biofilm-containing material; specificities were 98% for both. {Based on these findings, we are evaluating/developing a rapid PCR panel strategy for intraoperative use.} Despite the improved sensitivity of the approach developed in our prior funding period, there remain PJI cases that test negative. {In our renewal R01 application, we will focus on determining the etiology of culture- and panel PCR- negative PJI. We will also delineate subpopulations of same- and mixed-species biofilms associated with PJI, which our new preliminary data indicate are present and which may inform patient management, ensuring an ideal outcome. Finally, we will determine whether organisms not targeted by our PCR panel might account for what are currently classified as non-infectious arthroplasty failures unrelated to device failure.} We propose to use deep sequencing technology which has emerged and matured since our original R01 project period. Metagenomic and metatranscriptomic approaches will be used to analyze host and microbial nucleic acids in biofilm-containing materials dislodged from the surfaces of orthopedic implants. Using deep sequencing, we will analyze microbial DNA and total expressed RNA, the latter separated into microbial and human RNA. We will test the hypothesis that the new deep sequencing approaches will detect PJI with improved clinical sensitivity compared to culture and panel PCR when applied to biofilm-containing materials dislodged from surfaces of orthopedic implants. We will also define the prevalence of previously undetected polymicrobial PJI, and assess for subpopulations of same-type bacteria, which our preliminary data suggest will be present, and which can impact treatment as well as inform pathogenesis. We will analyze human gene expression in material dislodged from explanted arthroplasties to test the hypothesis that there are PJI-specific human genomic expression signatures. {The cause of aseptic failure is in many cases incompletely defined; several lines of evidence suggest that cases not caused by device-related failure may be caused by bacteria. To address this, we will use metagenomic and metatranscriptomic approaches to study aseptic failure, separately analyzing failure due to particulate wear debris, implant malposition/structural device-related failure and more strictly classified aseptic failure.} Beyond definition of the etiologies of PJI and aseptic failure, our results will inform development of new diagnostic assays for PJI targeting microbes directly and/or their associated host response and will provide targets for prevention (e.g., vaccination) and therapy (i.e., drug development) of PJI. In addition, we will determine whether some patients undergoing revision hip or knee arthroplasty for aseptic arthroplasty failure have previously unrecognized infection.